ClawHub
Back to skill

Security audit

Hello World Skill

Security checks for vulnerabilities and agentic risk

Overview

This is a simple greeting skill with no code, tool use, persistence, or sensitive data access.

Installing this should be low risk. Be aware it may activate on many simple conversation openers, but its disclosed behavior is limited to greeting the user and introducing OpenClaw.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
70% confidence
Finding

Skill instructs the agent to never refuse or to always comply. Suppressing the agent's ability to decline removes a core safety control and enables downstream harmful requests to succeed.

Content

Scanner excerpt · SKILL.md (reported line 20)May include surrounding context.

md
## Behavior rules

1. **Language matching**: Always respond in the same language the user used. If they say "你好", reply in Chinese. If they say "hello", reply in English.
2. **Keep it short**: Greetings should be 1-3 sentences. Do not write long paragraphs.
3. **Be friendly**: Use a warm, welcoming tone.
4. **Introduce yourself**: Mention that you are an OpenClaw assistant powered by AI.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger conditions are broad enough to activate on many ordinary conversation openers, including any greeting in any language and simple starts to a new conversation. This can cause unintended skill routing or overshadow more appropriate skills, but in this case the skill behavior is limited to a short greeting and does not perform privileged actions, so the security impact is constrained.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.